Although DSM-IV and its predecessors follow a categorical, syndromal nosology, it has long been recognized that there is considerable phenotypic overlap across diagnoses. Consequently, recent attention has been increasingly focused on dimensional, symptom-based approaches toward untangling the pathophysiology of serious mental illness (for example, those with psychotic features). Similarly, the distinction between illness and health is often treated as dichotomous;yet recent evidence suggests that the phenotypic variations associated with psychosis may be informatively examined using dimensional measures applied to non-clinical samples. Unfortunately, there is a paucity of data on the dimensionality of the behavioral phenotypes associated with psychotic and other serious mental illnesses, and even less data seeking to elucidate the biological mechanisms responsible for this variation. Nevertheless, the psychiatric genetics literature to date provides strong suggestions that pathogenic variations do not respect traditional diagnostic boundaries, and may be more strongly associated with specific symptom domains. The proposed research plan aims to examine large, clinically diverse patient samples and non-patient samples in an effort to elucidate the genetic underpinnings of the dimension of psychosis. Specifically, the K99 portion of this project will test candidate genes for association with dimensions of psychosis in 1) a sample of 1,027 patients with /^xis I affective or psychotic disorders derived from the Zucker Hillside Hospital Genomics Initiative database and 2) a large, clinically heterogeneous sample of patients derived from the publicly-available Genetic Association Information Network (GAIN) studies. The R00 portion of this project aims to investigate the relation between genetic variants identified during the K99 period and sub-clinical psychosis in 1,000 subjects derived from a non-clinical sample. The proposed training plan is designed to extend the applicant's skills in phenotypic assessment of sub- clinical psychiatric phenomena, as well as to provide systematic, advanced training in statistical genetics. The mentorship plan will include formal didactics, one-on-one tutorials, and hands-on experience in these areas, as well as additional training in molecular genetics and the responsible conduct of research, with the goal of independent submission of an ROI to test multivariate models of genotype-phenotype relationships across the psychotic spectrum. RELEVANCE (See instructions): The identification of genetic factors that underlie psychosis and its related phenotypes may substantially improve our understanding of the trajectories of psychiatric illnesses. This study seeks to elucidate the genetic underpinnings of the psychosis dimension in clinical and non-clinical samples to facilitate a better understanding of the etiology of psychiatric disorders.